See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/229441088

Porcelain Casting Slips Formulated with Waste Glass

Article  in  International Journal of Applied Ceramic Technology · July 2008

DOI: 10.1111/j.1744-7402.2008.02266.x

CITATIONS READS

6 65

2 authors:

S. R. Bragança Carlos P Bergmann

Universidade Federal do Rio Grande do Sul Universidade Federal do Rio Grande do Sul
78 PUBLICATIONS   424 CITATIONS    380 PUBLICATIONS   3,676 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Synthesis of metal oxides through biomimetização eggshell membranes View project

TiO2 thin films: Doping and functionalization for sensors applications View project

All content following this page was uploaded by S. R. Bragança on 03 October 2019.

The user has requested enhancement of the downloaded file.

 Int. J. Appl. Ceram. Technol., 6 [2] 264–269 (2009)
DOI:10.1111/j.1744-7402.2008.02266.x

Ceramic Product Development and Commercialization

Porcelain Casting Slips Formulated with Waste Glass

Saulo R. Bragança* and Carlos P. Bergmann
Department of Materials, Federal University of Rio Grande do Sul, 90035-190 PoA-RS, Brazil

In this work, it was investigated as to what happens when waste glass is introduced into the composition of a ceramic body
in an industrial slip-casting process. For pressed products, such as tiles and bricks, the use of waste glass is sufficiently simple,
but when used for slip-casting products, such as china and porcelains, the presence of glass destabilizes the slip. A better
stabilization of the casting slip was achieved with an organic deflocculant instead of sodium silicate. Therefore, the factors that
determine the success or the failure of a deflocculant were analyzed, such as pH, x potential, viscosity, and thixotropy.

Introduction firing temperature, besides the utilization of a recycled

material.
The composition of triaxial ceramics is 50% kaolin/ In products formed by pressing or extrusion, such
ball clay, 25% quartz, and 25% feldspar/nepheline sye- as bricks and tiles, the use of waste glass is quite simple:
nite. The proportions vary considerably, and sometimes washing, milling, and the appropriate mixing of the
other raw materials are present. Therefore, the compo- waste glass to the industrial batch, mainly replacing part
sition of diverse products, such as sanitarywares, china, of conventional fluxes. However, for ceramics products
artistic porcelains, and tiles, is based on these raw ma- formed by slip casting, the presence of waste glass gen-
terials. In this triaxial system, feldspar or nepheline sye- erates problems concerning the stabilization of the cast
nite acts as a fluxing agent, reducing the firing slip. The change in the batch rheological properties was
temperature of the ceramic parts and improving the pointed out by Jackson et al.2 Slip casting problems are
aesthetic and technical properties of the final pieces. probably related to the release of ions from waste glass
The advantages of using waste glass partially re- that are time dependent. This effect makes the control
placing feldspar or nepheline have been proved through of viscosity and slip thixotropy more complex.
many works, showing the application of waste glass in The surface charging behavior of a ceramic particle
red ceramics,1 sanitaryware and white porcelains,2,3 and suspended in water is complex, and depends basically on
porcelain stoneware4,5 compositions. Interesting aspects three mechanisms: desorption of ions from the particle
of microstructure investigation have been published re- surface, chemical reaction with the aqueous medium,
cently.6,7 These works pointed out important economic and adsorption of impurity ions or specific additives. At
advantages, for example, the reduction of the process’ a given pH, the surface charge depends on the valence
and population of ions in suspension, and ions bonded
*[email protected]
to the surface particle. Suspension stability is explained
r 2008 The American Ceramic Society by the DLVO theory, according to the development of
www.ceramics.org/ACT Porcelain Casting Slips with Waste Glass 265

an energy barrier to coagulation, keeping the system Table II. Characterization of the Slip as Received
deflocculated. An alternative method is the adsorption
Industrial
of a polyelectrolyte that can provide a steric hindrance
to prevent coagulation.8–10 Technical parameter slip
The aim of the present work was to investigate as to Density (g/cm3) 1.78
what happens when waste glass is introduced into the Temperature (1C) 31
composition of a ceramic body in an industrial slip- Moisture (%) 25
casting process, and why the addition of waste glass Minimum viscosity (cps) 778
makes the slip unstable for ceramic processing. Partic- Retained on sieve 325# (%) 3.09
ularly, a composition adjusted for production of sani- Particle size o10 mm (%) 70
tarywares was used, investigating changes in pH, x Particle size o1 mm (%) 22
potential, viscosity, and thixotropy. Organic material (%) 1.8
Sulfate (ppm) 105
Experimental Procedure Soluble salts (ppm) 515

The casting slip used in this work came from san-

itaryware production. The average chemical composi-
Waste glass from recycled milled bottles/plates was
tion and parameters of industrial slip characterization
added to the industrial casting slip and then the density
are shown in Tables I and II, respectively.
was adjusted to the proper value with addition of water.
The characterization of the industrial slip (without
Following this, viscosity (disk viscometer, Brookfield
waste glass addition) was made according to the method-
model RVDVII1), thixotropy (Gallenkamp torsion
ology adopted in the company, and the parameters are
viscometer, model GT 0888, Gabbrielli Technology,
described below: (a) Viscosity: digital viscometer (model
Calenzano, Italy), x potential (model Zeta Plus, Brook-
RVDVII1, spindle 4, rotation 30 rpm, Brookfield,
haven Instruments, Holtsville, NY), and pH were
Stoughton, MA). The deflocculation curve was measured
measured. Rheological properties were evaluated con-
with samples of 500 cm3. The deflocculant agent (sodium
sidering the addition of sodium silicate (B61% sodium
silicate/organic deflocculant) was added until the mini-
silicate concentration) or an organic deflocculant named
mum viscosity was reached. (b) Density and material on
Ecodefloc AD-4 (electrosteric and anionic deflocculant,
sieve#325: density was measured with the aid of an
a modified polymer from plant extraction bioflavonoid,
analytical scale in samples of 100 cm3. Afterwards, the
short polymer chain) produced by Acquaquimica, Novo
amount of material retained by screen mesh 325
Hamburgo, Brazil. The deflocculation agents were com-
(445 mm) was measured. (c) Particle size: particle size
pared in terms of their effect on the slip stabilization.
was measured in a laser particle size analyzer (CILAS
Two kinds of casting slips were used. The first one,
model 1180). (d) Sulfate/soluble salts: spectrophotometer
named aged slip, was allowed to age for a month. The
(model LAMBDA L650, PerkinElmer, Waltham, MA).
second slip, named new slip, was a new one (allowed to
age for a day), almost ready for use. The waste glass was
introduced 4 h before running any test. The particle size
Table I. Chemical Analysis
distribution of waste glass was similar to feldspar (90%,
Industrial slip Waste glass o47 mm; 50%, o30 mm; 10%, o4 mm; 2%, o1 mm).
The choice of using 5% waste glass in the batch is ex-
SiO2 68.4 73.0 pected not to interfere significantly with the batch phys-
Al2O3 24.2 2.4 ical properties.
K2 O 3.4 0.3 Before recording the data to obtain the deflocculation
Na2O 1.8 14.1 curves in the Brookfield viscometer, the slip was manually
Fe2O3 0.7 o0.1 and gently stirred for 1 min. After each deflocculant agent
TiO2 0.3 o0.1 addition, the slip was stirred again and then the data were
MgO 0.3 1.6 recorded. For thixotropic analysis, the slip was previously
CaO 0.25 8.2 deflocculated using sodium silicate.
266 International Journal of Applied Ceramic Technology—Bragança and Bergmann Vol. 6, No. 2, 2009

Results glass structure, such as calcium.11 This hypothesis was

confirmed by flame photometer analysis using milled
The influence of the waste glass addition on pH was glass in water. This test was performed using a concen-
analyzed firstly in water and then in slip. This effect was tration of glass of 0.04 g/L in water, stirred for 10 min.
compared with the use of feldspar. Secondly, the effect The concentration of solubilized ions Na and Ca in
of waste glass on the x potential was evaluated. Finally, water was 0.11 and 0.004 wt%, respectively.
the curves of viscosity versus amount of deflocculant La Course and Manson12 proposed that the leach-
added were compared, considering the use of an organic ing of glass ions, alkali and alkaline earth ions, leads to
deflocculant and conventional sodium silicate. The an instantaneous increase in pH according to the fol-
thixotropic behavior and the characterization of the lowing reaction:
cast pieces were analyzed under the ideal deflocculation
condition for each slip. ðSiO... Rþ Þglass þ H2 Osol

! SiOHglass þ Rþ
sol þ OH ð1Þ
pH evaluation
At high pH (pH410), dissolution of the silicate
Table III shows that the addition of waste glass or network may occur.12 In the present work, the dissolu-
feldspar had an alkalinizing effect on the pH value of tion of silicate network ions was not expected to occur,
deionized water (after 5 min of stirring). Waste glass had because the pH was approximately 8, using 10% of glass
a stronger effect than feldspar. However, in the test us- in ceramic slip casting, according to Table III. In an-
ing the slip, the alkalinizing effect was lower. In this other experiment, the dissolution of divalent cations was
situation, feldspar modifies the initial pH slightly. Con- minimal for nepheline syenite, but significant for mo-
sequently, it can be inferred that the release of ions from novalent cations.13
feldspar is low at this concentration. On the other hand, As a consequence of ion leaching from glass parti-
the effect of glass in a slip is more significant, leading to cles, the control of pH is complex and stabilization of
an increase of pH from 6.5 to 8.1, on addition of 10% the slip is difficult, at least for a short period of time. It
waste glass. will depend on the glass content and particle size dis-
The alteration of pH caused by addition of glass in tributions, besides temperature, mixing time, and so on.
water or slip is due to the quick release of ions from The instant changing of pH when glass is added
waste glass leading to hydroxide ions&apos formation. may be one of the causes for the suspension instability of
Using milled soda–lime glass, it is expected that sodium casting slip formulated with waste glass. In an experi-
and calcium will leach from glass particles. However, mental test, considering measurements directly in the
from glass science, it is known that the release of the slip 15% of waste glass, the pH remained approxi-
sodium ion is much easier than a stabilizing ion of the mately constant at B8 for at least 30 h. This means that
the time for mixing and then to age the slip is a key
factor to stabilize the slip.
Table III. Comparison of the Influence of Glass or
Feldspar Addition (%) on pH Measurements in
n Potential
Deionized Water or in the Industrial Slip (New Slip)
After 5 min of Stirring The x potential curve in Fig. 1 shows the behavior
PH of the aged slip and the new slip with 5% of waste glass.
Amount of glass By considering that a x potential higher than about
or feldspar 0% 1% 5% 10% 25 mV is necessary for deflocculation of a colloidal
slip,14 it can be pointed out that: (i) the aged slip
Water with or without waste glass addition is easily defloccu-
Glass 6.0 8.0 9.5 9.9 lated by adjusting pH and the increase in pH contin-
Feldspar 6.0 6.2 6.8 7.6 uously leads to a higher x potential in the range
New slip analyzed; (ii) the presence of waste glass markedly
Glass 6.5 6.8 7.5 8.1 changes the behavior of the new slip, making the slip
Feldspar 6.5 6.6 6.7 6.8 zeta potential unstable, even when the deflocculants
www.ceramics.org/ACT Porcelain Casting Slips with Waste Glass 267

new slip 3000

new slip + 5% waste glass Slip + 5% waste glass

Apparent viscosity (cps)

new slip + 5% waste glass + Na2SiO3 2500 Slip + Na2SiO3
new slip + 5% waste glass + organic def.
2000
0 aged slip
–10 aged slip + 5% waste glass 1500
Zeta potential (mV)

–20 1000
–30
500
–40
–50 0
0 10 20 30 40 50 60 70 80
–60 Reaction time (min) or % Na2SiO3(x10–3)
–70
4 5 6 7 8 9 10 11
pH Fig. 2. Viscosity as a function of reaction time of 5% waste glass
added to the aged slip, and deflocculation curve of the aged slip with
Fig. 1. x potential versus pH. The suspension concentration is sodium silicate (% Na2SiO3 dry-weight basis).
0.001 solid volume/liquid volume in deionized water. pH is
adjusted using NaOH or HCl. The deflocculant concentration is
0.01% on a dry-weight basis. previous assay, the viscosity of the industrial slip (with-
out waste glass) was about 2500 cps, and this value did
not change during the assay (60 min). Also in Fig. 2, it
were added—standard deviations increased from 76% can be seen that the deflocculating effect of sodium sil-
to 720%; (iii) sodium silicate (Na2SiO3) has an effi- icate is similar to the glass addition.
ciency similar to NaOH addition for the new slip with
waste glass addition, for an adjusted pHB8; (iv) a good Comparison between the Deflocculation of the
x potential of 4074 mV for pHB7.2 was obtained Industrial Slip with Glass Using Sodium Silicate
for the new slip with waste glass addition, using the or- or an Organic Deflocculant
ganic deflocculant.
As a consequence of the waste glass slip instability, The deflocculation curves of the aged slip with
the results indicated in Fig. 1 have to be confirmed in waste glass, considering both sodium silicate and or-
industrial tests.15 ganic deflocculant additions, are shown in Fig. 3. It can
be seen that just a narrow deflocculation range was ob-
Effect of Waste Glass on Slip Viscosity tained with the use of sodium silicate. This proves the
influence of a higher concentration of ions, after aging
The addition of waste glass in a slip causes a de- for a month. As a consequence of this, it did not have
crease in the viscosity value, which is a very interesting the same efficiency as the organic deflocculant, which
result and is also beneficial for the slip-casting process. can yield a lower viscosity and a higher x potential as
This can be a consequence of a higher pH, as explained shown in Fig. 1, probably due to the steric repulsion
in equation 1. It can also be pointed out here that the mechanism.
interaction with kaolin particle may be similar to the In the deflocculation curve of the new slip shown in
behavior of the sodium silicate deflocculant (water glass) Fig. 3, a reduction of viscosity with addition of glass can
in a traditional ceramic slip. The release of Na1 pro- also be noticed. But, in this case, both the organic
motes an ion-exchange mechanism, removing the un- deflocculant and the sodium silicate reached a suffi-
wanted calcium (Ca21), and consequently increasing ciently low viscosity. However, the organic deflocculant
the x potential.8,9 Besides, the adsorption of OH or continues to reduce viscosity, while a trend toward an
SiO23 increases the negative charge of kaolinite parti- increase in viscosity with increasing amount of
cles, increasing repulsion forces.13 deflocculant can be observed for sodium silicate; fur-
Figure 2 shows that the slip viscosity starts to de- ther addition leads to a higher ion concentration, which
crease just after addition of waste glass (time zero is the means a lower x potential.14 As it is well known, wider
instant when waste glass was added to the slip). In a ranges of deflocculation in relation to the amount of
268 International Journal of Applied Ceramic Technology—Bragança and Bergmann Vol. 6, No. 2, 2009

aged slip + 5% waste glass + Na2SiO3 were taken in a Gallenkamp viscometer. The thixotropic
aged slip + 5% waste glass + org. deflocc.
new slip + org. deflocc.
analysis was evaluated in terms of the difference in de-
new slip + 5% waste glass + org. deflocc. grees between the readings of the first and the second
1400 new slip + 5% waste glass + Na2SiO3 measurements. These data provide the thixotropic be-
Apparent viscosity (cps)

1200 havior of the slip for ceramic parts’ casting. The results
1000 can be found in Table IV.
800 In Table IV, it can be seen that the new slip defloc-
600 culated with the organic deflocculant achieved the de-
400 sired rheological properties. However, the slip
200 deflocculated with sodium silicate was unstable accord-
0 ing to the test performed using a Gallenkamp viscom-
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 eter, with a significant variation of values between the
% deflocculant
two readings. Using the slip1sodium silicate, it was
Fig. 3. Deflocculation curves with sodium silicate or an organic
impossible to produce casting parts with conditions
deflocculant (% dry weight basis). Aged slip15% waste glass, new similar to the tests performed on a day-to-day basis in
slip, and new slip15% waste glass. the laboratory of the company. On the other hand, the
slip1organic deflocculant produced good casting parts
with a smooth surface and easy detachment from the
deflocculant added improve the cast processing, because plaster mold. This proved the viability of using glass in
this means that a more stable slip is cast, and particle the composition of an industrial slip.
precipitation is more difficult to occur. This is a signifi-
cant advantage shown by the organic deflocculant.
Discussion

Thixotropy Behavior The glass addition in a slip results in the formation

of hydroxide ion, increasing the pH of the mixture sig-
In this item, the thixotropy of the slip was evalu- nificantly (Table III). The adsorption of OH on kaoli-
ated. The same laboratory routine that is used in the nite particles increases the deflocculation of the slip,
industry was adopted, following the procedure of a because of an increase in the repulsion strength between
commercial slip for sanitaryware production. The ini- particles. This is the double layer or the DLVO theory,
tial fluidity was measured after 1 min of manual stirring. which is well discussed in the literature.14,16 Because
The second fluidity measurement was taken after the glass promotes deflocculation, it is a fact that the release
slip was allowed to stand for 6 min; both measurements of sodium superposes the release of calcium from the

Table IV. Comparison of Slip Fluidity Data (New Slip) Using a Gallenkamp Viscometer between the Use of
Sodium Silicate and an Organic Deflocculant
Deflocculated Slip 1 Deflocculated slip 1
Slip 1 5% glass 5% glass
Slip 5% glass (organic deflocculant) (sodium silicate)
Quantity of deflocculant   0.031% 0.062% 0.028% 0.056%
Density (g/mL) 1.78 1.77 1.77 1.77 1.77 1.77
Brookfield viscosity (cps) 1240 1010 525 460 648 663
Initial fluidity (degree) 167 215 251 255 236 225
Fluidity after stand for 6 min (degree) 13 161 228 246 194 199
D degree (Thixotropy) 180 54 23 9 40 26
Temperature (1C) 23 23 23 23 23 23
Sodium silicate was used in milling, clay blunging, and batch mixing, but with no further addition.
www.ceramics.org/ACT Porcelain Casting Slips with Waste Glass 269

glass. This was confirmed by flame photometer analysis. which guarantees a steady interval for the slip-casting
Therefore, the results in Fig. 2 confirm that addition of process.
glass promotes the reduction of the viscosity of the slip.
Glass leaching is a time-dependent process, pro- References
moting a slow but strong change in viscosity (Fig. 2)
and an unstable x potential (Fig. 1), which explains the 1. A. S. Smith, ‘‘Recycled Glass as a Brick Fluxing Agent,’’ Proceedings of Sus-
difficulty in stabilizing the slip. This is in accordance tainable Waste Management and Recycling: Glass Waste. eds., M. C. Lymba-
chiya and J. J. Roberts. Thomas Telford, London, 149–165, 2004.
with the results of the use of sodium hydroxide as a 2. P. R. Jackson, P. Hancock, and D. Cartlidge, ‘‘Recycled Bottle Glass as a
deflocculating agent, which requires a long time for sta- Component of Ceramic Sanitaryware,’’ Proceedings of Sustainable Waste Man-
agement and Recycling: Glass Waste. eds., M. C. Lymbachiya and J. J. Roberts.
bilization of the slip.17 Hence, the introduction of a Thomas Telford, London, 313–320, 2004.
glass powder with a large particle size distribution results 3. S. R. Bragan@a and C. P. Bergmann, ‘‘Use of Waste Glass in Porcelain,’’ Pro-
ceedings of Sustainable Waste Management and Recycling: Glass Waste. eds., M. C.
in extra difficulties in the stabilization of the slip. Lymbachiya and J. J. Roberts. Thomas Telford, London, 217–224, 2004.
The addition of the organic deflocculant propitiates 4. F. Matteucci, M. Dondi, and G. Guarini, ‘‘Effect of Soda–Lime Glass on
the decrease of slip viscosity by the creation of a physical Sintering and Technological Properties of Porcelain Stoneware Tiles,’’ Ceram.
Int., 28 873–880 (2002).
barrier, called a steric hindrance mechanism.14,16 This 5. A. Tucci, L. Esposito, E. Rastelli, C. Palmonari, and E. Rambaldi, ‘‘Use of
deflocculation mechanism is more stable and allows the Soda–Lime Scrap-Glass as a Fluxing Agent in a Porcelain Stoneware Tile
Mix,’’ J. Eur. Ceram. Soc., 24 83–92 (2004).
use of glass powder in a ceramic slip, according to the 6. S. R. Bragan@a and C. P. Bergmann, ‘‘Traditional and Glass Powder Porce-
results shown in Table IV. The development of organic lain: Technical and Microstructure Analysis,’’ J. Eur. Ceram. Soc., 24 2383–
2388 (2004).
deflocculants with competitive prices in replacement of 7. T. Tarvornpanich, G. P. Souza, and W. E. Lee, ‘‘Microstructural Evolution
sodium silicate, such as that used in this work, allows on Firing Soda–Lime–Silica Glass Fluxed Whitewares,’’ J. Am. Ceram. Soc.,
88 [5] 1302–1308 (2005).
the utilization of glass without increasing the cost of 8. W. E. Worral, ‘‘The Colloid Chemistry of Clays and Ceramics Materials,’’
deflocculant utilization. Ceramics Monographs—Handbook of Ceramics, Verlag Schmidt, Freiburg,
Germany, 1–5, 1982.
9. R. A. McCauley, ‘‘Rheology,’’ Ceramics Monographs—Handbook of Ceramics,
Verlag Schmidt, Freiburg, Germany, 1–7, 1983.
10. G. W. Phelps, ‘‘Slip Casting,’’ Ceramics Monographs—Handbook of Ceramics,
Conclusions Verlag Schmidt, Freiburg, Germany, 1–9, 1982.
11. J. M. F. Navarro, ‘‘El vidrio (Glass),’’ Consejo Superior de Investigaciones
Cientı́ficas, Madrid, 1991.
1 21
Waste glass releases Na , Ca , and results in the 12. W. C. LaCourse and W. Manson, ‘‘Glaze Problems From a Glass Science
creation of OH in the slip. This causes an increase in Perspective,’’ Science of Whitewares. eds., V. E. Henkes, G. Y. Onoda, and
W. M. Carty. American Ceramic Society, Westerville, OH, 339–356, 1996.
pH over time, promoting the deflocculation of the 13. M. W. Carty and U. Senapati, ‘‘Porcelain–Raw materials, processing, phase
slip. This makes the stabilization of the rheological evolution, and mechanical behavior,’’ J. Am. Ceram. Soc., 81 [1] 3–20 (1998).
14. J. S. Reed, Principles of Ceramic Processing, 2nd Edition, John Wiley & Sons,
properties (of the slip) more difficult, and as it is well Oxford, U.K., 1995.
15. M. W. Carty, K. R. Rossington, and U. Senapati, ‘‘A Critical Review of
known, the control of these properties is fundamental to Dispersants for Whitewares Applications,’’ Science of Whitewares II. eds.,
the slip-casting process. The use of the organic W. M. Carty and W. Sinton. American Ceramic Society, Westerville, OH,
deflocculant allowed stabilization of the slip, which 155–167, 2000.
16. C. H. Schilling and I. A. Aksay, ‘‘Slip Casting,’’ ASTM Engineered Materials
was not accomplished with sodium silicate, as shown Handbook Vol. 4: Ceramics and Glasses. ASTM International, Cleveland, OH,
153–158, 1991.
in this work. Measurements of viscosity as a function of 17. R. Herrmann, ‘‘Slip Casting in Practice,’’ Ceramics Monographs—Handbook
deflocculant additions showed a large viscosity plateau, of Ceramics, Verlag Schmidt, Freiburg, Germany, 1–5, 1989.

View publication stats